Cognitive Load Theory: An Overview
University of South Florida
As much as education has changed in the last 100 years, many things have stayed the same. Many of the technologies we now use are just electronic analogs of real world materials. Powerpoint is an electronic chalk board, e-books are more functional textbooks. But because of the ease of creating media in these types of software, we often use techniques because they are expedient, not because they’re necessarily educationally sound. Many times the software itself, because of templates or layout, channels us into using unsound techniques. Even though students might prefer Powerpoint over methods of delivery like whiteboards, or transparencies, research has shown they don’t necessarily retain information better (Cooper, 2009).
Richard Mayer quantified the Cognitive Theory of Multimedia as “meaningful learning (that) occurs when learning engages in appropriate verbal and visuospatial thinking” (Mayer, 2002, p. 61). According to Mayer’s theory, the two main channels to taking in information are the eyes and ears. Most of what we learn comes from seeing and hearing. We take that information, put it into working memory and monitor, integrate, calculate, or otherwise manipulate that data, which then becomes stored in long term memory. Sweller referred to the amount of effort needed to manipulate that data as cognitive load (Sweller, 1988).
According to Sweller’s theory, there are two main types of cognitive load, intrinsic, and extraneous. Intrinsic load is the inherent complexity of the material being learned. Chase and Simon found that the difficulty can be mitigated by learners with higher levels of expertise. For example, a chess master can remember more moves than a novice. The expert chess player has developed schema to manage their learning and recall, reducing their cognitive load compared to a novice. Their study involved players reproducing certain positions after viewing them for 5 seconds. The expert players were able to reproduce the board with more accuracy. Interestingly enough, when the pieces were placed randomly, the experts had no advantage over the novices. (Chase, W.G. & Simon, H.A., 1973)
Extraneous load is introduced by the ways materials are presented, which complicates the learning task. If text and narration are presented simultaneously, then the visual and auditory channels are overloaded, decreasing learning. Recently, Sweller has proposed that there is a third type of load, germane load, which is the load manipulated by the instructional designer, analogous to Vygotsky’s proximal zone of development. According to Kalyuga et. al, intrinsic load is not fixed, but can be manipulated (expert schema, as in the chess example above), and that intrinsic load can create “good” effort (again, think Vygotsky’s zone of proximal development), and is indistinguishable from germane load. Kalyuga posits that germane load is an unnecessary complication, was introduced on a theoretical basis, and is unsupported by empirical data.
Each side has it’s followers, and the debate is ongoing, but whether you subscribe to Kalyuga’s or Sweller’s view, the important thing is that the instructional designer reduce extraneous cognitive load as much as possible. The intrinsic load (or germane load, if you wish) can be manipulated by the designer to match a students level, and achieve the learning goals desired, but any extra load introduced can impede learning.
Cognitive load theory has implications not only for designers of online instruction, but for classroom teachers too. The way material is presented in face to face teaching, either by print or computer (powerpoint) can hinder, or enhance learning. Taki and Aksu found that designing class materials using Richard Mayer’s cognitive load reduction principals could increase scores on middle school students algebra tests (Taki, A. and Aksu, M., 2012). It’s also important to note that reducing cognitive load might not automatically increase learning. Yang found that attention cueing (signaling principal) using a model of the human circulatory system could reduce cognitive load but did not increase scores on tests. (Yang, H. Y., 2016) So while reducing cognitive load is good, it might not automatically translate to higher test scores.
How is cognitive load measured? Typically cognitive load is measured by a self reported Likert rating applied after each task, then averaged, or after all tasks have been completed. Sweller found these reports to be one of the most sensitive measures of cognitive load.
Schmeck et. al., found that when measured at the end of a series of tasks, cognitive loads were self reported as being higher, than when they were reported after each sub-task. They suggested that some of the newer psychometric measures, such as eye blink, or heart rate monitoring might be more accurate measures. (Schmeck, A. a., Opfermann, M., Gog, T., Paas, F., & Leutner, D., 2015). Considering this, it’s important when reading cognitive load research to pay specific attention to the measurement instruments, as different instruments may produce different results.
So how can we reduce cognitive load while designing our instruction? Mayer made 3 theory based assumptions to keep in mind when designing instructional materials.
Dual channel assumption: This theory takes into account the two parts of working memory, the visual and the auditory channel. If two sources of visual information are included (text and graphics) then the mind must process those simultaneously, which increase the chance of cognitive overload.
Limited channel assumption: The human brain has a limited capacity to take in information on each channel. This capacity can be vary by person, or by training as shown in the chess example above. When more the brain is taking in more information than it can process, overloading occurs.
Active processing assumption: This theory takes into account the way humans process information. Meaningful learning occurs when learners actively process audio and visual information. (Mayer, 2002)
In addition to these assumptions, Mayer conducted studies to determine what were the best ways to present text, graphic, audio, and audiovisual information. From these studies he developed seven principals that could be applied to instructional design. These principles are simple, practical, and are easily integrated into lesson planning and design. Designers of Power Point presentations would do well to keep these principles in mind.
Multimedia Principle: Students learn better from multimedia presentations than simply text or audio presentations.
Contiguity Principle: Students learn better when narration and pictures are presented simultaneously, rather than consecutively.
Coherence Principle: Students learn better when the multimedia presentation is expanded and interesting, rather than basic. One caution is the too many unnecessary or irrelevant sounds or images can decrease learning.
Modality Principle: Animation and narration are more effective than animation and onscreen text.
Personalization Principle: Learning is enhanced when words are presented in a conversational rather than a formal style.
Interactivity Principle: Learning is enhanced when students can control the rate of presentation. Signaling Principle: Learning is enhance when students attention is drawn to important details (highlighting for example). (Mayer, 2002)
In addition Park found that using onscreen avatars (human faces) to present narration was even more effective than narration alone. Faces can serve as effective signaling cues (signaling principle), but this research shows that it may also increase user confidence and perceptions of relevance. (Park, S., 2015)
Whatever the format or delivery methods using these principles can help the instructional designer or classroom teacher increase learning in their students. Cognitive load theory, as the name implies is still just a theory, and research is ongoing. Keep in mind when reading research what instruments are used to measure the load, and that reduction of cognitive load might not automatically translate into long term learning gains. However Sweller, Mayer, Kalyuga, and others have generated enough empirical results that we can be relatively confident in turning their research into practice, and at the very least we are doing no harm.
Chase, W.G. & Simon, H.A. (1973). “Perception in chess”. Cognitive Psychology 4 (1): 55–81. doi:10.1016/0010-0285(73)90004-2.
Cooper, E. (2009). Overloading on Slides: Cognitive Load Theory and Microsoft’s Slide Program PowerPoint. AACE Journal, 17(2), 127-135.
Kalyuga, S. s. (2011). Cognitive Load Theory: How Many Types of Load Does It Really Need?. Educational Psychology Review, 23(1), 1-19.
Mayer, r.E. (2002). Cognitive theory and the design of multimedia instruction: an example of the two-way street between cognition and instruction. New Directions in Teaching and Learning, 89, 55-71.
Park, Sanghoon. (2015). The Effects of Social Cue Principles on Cognitive Load, Situational Interest, Motivation, and Achievement in Pedagogical Agent Multimedia Learning. Journal of Educational Technology & Society, 18(4), 211–229. Retrieved from http://www.jstor.org/stable/ jeductechsoci.18.4.211
Schmeck, A. a., Opfermann, M., Gog, T., Paas, F., & Leutner, D. (2015). Measuring cognitive load with subjective rating scales during problem solving: differences between immediate and delayed ratings. Instructional Science, 43(1), 93-114.
Sweller, J. (1988). Cognitive load during problem solving: Effects on learning. Cognitive Science, 12(2), 257-285. doi:10.1207/s15516709cog1202_4
Takir, A., & Aksu, M. (2012). The effect of an instruction designed by cognitive load theory principles on 7th grade students’ achievement in Algebra topics and cognitive load. Creative Education, (2), 232.
Yang, H. Y. (2016). The Effects of Attention Cueing on Visualizers’ Multimedia Learning. Educational Technology & Society, 19 (1), 249–262.